© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 2202
    Prev
    Next
    INDEX
    Explanations

    kind of / sort of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     каж
    -0.09
    æ¬²æľĽ
    -0.09
    æµĶ
    -0.09
    è·¯ç¨ĭ
    -0.08
    ams
    -0.08
    贯彻èIJ½å®ŀ
    -0.08
     vir
    -0.07
    æīĵäºĨ
    -0.07
    stud
    -0.07
    ä¸įæŃ¢
    -0.07
    POSITIVE LOGITS
    æĢ§æł¼
    0.10
    avit
    0.10
    /how
    0.10
    ä¼ļéĢīæĭ©
    0.10
    ç³Łç³ķ
    0.10
    ondo
    0.09
    ä¼ļ产çĶŁ
    0.09
    esser
    0.09
    èĬ¥
    0.09
    é¢Ŀå¤ĸ
    0.09
    Activations Density 0.012%

    No Known Activations